Find X for a given Y using interpolation

Hi all,

Suppose

x = [5 8 9 11 12.5 17];
y = [2 3.5 4.0 5.3 6.7 8.2];

Now I want to find the corresponding value for x where y = 4.8 using interpolation. I was wondering if there is any function in MATLAB which can do this calculation.

Answer 1

[Dyuman Joshi]

x and y are not mapped one-to-one for interpolation, there is an element missing in y.

interp1 is what you are looking for.

[Roohollah]

Thanks for reminding me for the issue.

What I need is a function to input x,y and y = 4.8. Then it returns the corresponding x for y = 4.8.

[Roohollah]

Many thanks.

interp1 worked well.

[Dyuman Joshi]

Look into the documentation (the link I attached above) for more info.

x = [5 8 9 11 12.5 17];
y = [2 3.5 4.0 5.3 6.7 8.2];

x0 = interp1(y,x,4.8)

x0 = 10.2308

[Roohollah]

Seems to be wrong. x0 must be between 11 and 12.5.

[Dyuman Joshi]

As y0 is between 4 and 5.3, so correspondingly, x0 should be between 9 and 11, which it is.

[Roohollah]

Yes, you are right.

As the last question. I am working on my real data. This is the code I use;

delta40 = interp1(F,delta,F40);   

And I get the following error message:

Error using matlab.internal.math.interp1

Sample points must be unique.

Error in interp1 (line 188)

VqLite = matlab.internal.math.interp1(X,V,method,method,Xqcol);

Error in Test (line 40)

F40 = 0.4 * Fmax; delta40 = interp1(F,delta,F40);

What does "Sample points must be unique." mean?

[Dyuman Joshi]

"What does "Sample points must be unique." mean?"

Let's introduce a repeating value in the data set -

y = [2 2 3.5 4.0 5.3 6.7 8.2 ];
x = [5 6.5 8 9 11 12.5 17 ];

Now if you want to interpolate for y between 2 and 3.5, say 3, which x values should be used for interpolation?

5-8 or 6.5-8? Interpolation for such a case does not make sense.

So, we have to use non-repeating values - a.k.a unique values in MATLAB terminology. That's what the error is stating.

Now, as 6.5-8 is a subset of the range 5-8 for the same value of 2, we can eliminate it from our data and then interpolate -

%Get unique values of the array in sorted manner
[y,idx] = unique(y)

y = 1×6

2.0000 3.5000 4.0000 5.3000 6.7000 8.2000

idx = 6×1

1 3 4 5 6 7

x = x(idx)

x = 1×6

5.0000 8.0000 9.0000 11.0000 12.5000 17.0000

interp1(y,x,3)

ans = 7

Note that this is just one of the possible cases of data with repeating values.

[Roohollah]

Many thanks